Patients with chemotherapy-refractory Hodgkin's disease (HD) have few treatment options. Approximately 40 percent of all cases of HD have been shown to be associated with Epstein Barr virus (EBV), characterized by a type II Latency pattern of infection and expression of fewer EBV antigens than in Latency III tumors. Previous studies have established that adoptive immunotherapy for EBV-induced lymphoproliferations in stem cell transplant and organ transplant patients (Latency III infections) with EBV-specific cytotoxic T lymphocytes (CTL) can lead to remission of disease. It is possible that similar strategies would be successful for Latency type II disorders. The objective of this study is to examine the clinical and immunologic effects of infusing donor-derived, EBVspecific CTL from HLA identical or haplo-identical donors for patients with relapsed/refractory, EBV-positive HD. The spectrum of the EBV antigens recognized by the CTL preparation and from patient T cells post-infusion will be determined, as well as levels of EBV specific CTL precursors (CTLp) by limiting dilution analysis (LDA). The therapeutic outcome will be assessed with clinical and radiographic endpoints and will be correlated with the level of donor/host HLA disparity, CTL reactivity against Latency II antigens, and levels of EBV CTLp post-infusion. We will track the infused CTL using PCR assays for short tandem repeats (STR). While the initial group of patients will receive EBV CTL without prior immunosuppression, subsequent cohorts of patients will receive a single CTL infusion following fludarabine, which will be used as an immunosuppressive agent to facilitate lymphoid engraftment. Since immunosuppression may increase the risk of graft vs. host disease, the CTL infusates and patient blood specimens will be examined for the presence of donor-derived, recipient specific T cells by LDA. To assess risk for CTL rejection, LDA will also be performed on post-infusion blood specimens to detect host-derived CD4 and CD8 cells with reactivity against donor antigens. This study will provide information on 1) whether therapy with allogeneic EBV specific CTL has clinical efficacy against EBV-positive HD; 2) if effective, whether this is correlated with effector cells specific to EBV latency type II antigens, and 3) if not effective, whether the failure is associated with a short half-life of the infused CTL.